Object detection model optimization method and device, equipment and storage medium
A technology of object detection and optimization method, which is applied in the field of image recognition, can solve problems such as misjudgment, achieve the effect of avoiding interference, solving misjudgment problems, and ensuring integrity
Pending Publication Date: 2021-10-12
CHINA MOBILE SHANGHAI ICT CO LTD +1
View PDF0 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0005] Embodiments of the present invention provide an object detection model optimization method, device, equipment, and storage medium objects, which realize pixel-level area detection, solve the problem of misjudgment caused by mutual occlusion of objects, and ensure the integrity of hats and objects , to avoid interference from occluded objects
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0081] The characteristics and exemplary embodiments of various aspects of the present invention will be described in detail below. In order to make the purpose, technical solutions and advantages of the present invention more clear, the present invention will be further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described here are only configured to explain the present invention, not to limit the present invention. It will be apparent to one skilled in the art that the present invention may be practiced without some of these specific details. The following description of the embodiments is only to provide a better understanding of the present invention by showing examples of the present invention.
[0082] It should be noted that in this article, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and ...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More PUM
Login to View More Abstract
The embodiment of the invention discloses an optimization method and device of an object detection model, equipment and a storage medium. The method comprises the steps of obtaining a data sample; according to the image data of the object, through an initial neural network model, generating a candidate region set including a to-be-detected region of the object; according to the candidate region set and the image data, performing segmentation through an initial neural network model to obtain a pixel-level contour detection region of the object; and according to the pixel-level contour detection area of the object and the identification information of the object, adjusting model parameters of the initial neural network model to obtain an optimized model. According to the optimization method and device of the object detection model, the equipment and the storage medium, the candidate region set is generated through the neural network model, and the full convolutional neural segmentation is performed on the image data of the object to realize the detection of the pixel-level target region of the object, so that the integrity of the object and the hat is ensured, the problem of wrong detection caused by overhigh overlapping rate between objects due to shielding is solved.
Description
technical field [0001] The invention relates to the field of image recognition, in particular to an object detection model optimization method, device, equipment and storage medium object. Background technique [0002] In existing object helmet detection methods, feature extraction, classifier training, and overlap rate calculation are often used to judge whether an object wears a helmet or not. [0003] However, when judging by using the overlap rate, when the body of the object and the object occlude each other, whether the occluded object wears a helmet or not, it will be wrongly judged as wearing a helmet based on the excessively high overlap rate. [0004] Therefore, when objects are mutually occluded, there will be a problem of incorrect matching and misrecognition of adjacent object helmets. Contents of the invention [0005] Embodiments of the present invention provide an object detection model optimization method, device, equipment, and storage medium objects, wh...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More Application Information
Patent Timeline
Login to View More Patent Type & Authority Applications(China)
IPC IPC(8): G06K9/00G06K9/34G06K9/62G06N3/04G06N3/08
CPCG06N3/084G06N3/045G06F18/241
Inventor 王赛楠祝晓旦王华飞
Owner CHINA MOBILE SHANGHAI ICT CO LTD